The overall goal of this interdisciplinary project is to research and develop novel methods for extracting information from scanned artworks and visualizing them such that art historians can gain further insights into those artworks and art forms.
The particular context will be ornamental artworks of the 18th century, the so-called rocaille, which provide a very rich set of shapes, structures, and, thus, challenging problems in computer science, especially machine learning. In this project, we will focus on etchings, which have already been scanned in high resolution.
The successful candidate will be closely working with art historians to define the appropriate representations, structures, and similarity measures to be extracted from the scanned artworks. Also, one important output of the project is to be a set of software tools that art historians can use later on in their work. These should allow them to classify rocaille ornaments, establish chains of evolutions of designs, distinguish between different authors, determine similarities among different artworks, etc.
Methods that will be needed for these tasks include machine learning, computer vision, graph algorithms, geometric computing, and good programming skills.
Applications should comprise a cover letter, complete CV including any achievements, degree certificates (including list of courses and grades), names and contact details of at least two referees, and other credentials if any (e.g., recommendation letters, publications, etc.). Application by email is preferred. We encourage candidates to apply immediately, but we will continue to review applications until the position is filled. If you have any questions about the position, please do not hesitate to address them to the above email address.
University of Bremen, Dept. of Computer Science, Bibliothekstr. 5, 28359 Bremen / Germany
zach at cs.uni-bremen.de